The evaluation of retinal structure as determined by photographic documentation at the three-month follow-up examination has indicated a favorable outcome in eyes treated in the CRYO-ROP trial. This outcome makes long-term follow-up of infants in the CRY0-ROP trial essential to evaluate the long-term safety and efficacy of the procedure, to determine the optimal disease threshold for application of the therapy, and to define the indications for routine follow-up of infants with mild retinopathy of prematurity. The primary outcome measure of the CRYO-ROP study was the status of the retinal structure, as determined by photographic documentation at one year after cryotherapy. During the time since the intake of patients for the trial began in 1986 the technology for measuring visual function in infants has evolved to the point that it has been possible to add a second outcome measure at one year after cryotherapy. This second outcome measure is grating visual acuity, as assessed monocularly by the Teller Acuity Card method. We propose to continue to evaluate retinal structure and visual function in 207 infants in the CRYO-ROP trial for an additional five-year follow-up period in order to attempt for the first time to correlate the structural changes observed in both the acute proliferative and scarring phases of ROP with the eventual visual function of the eye. We will also evaluate the long-term risk/benefit ratio of recommending cryotherapy in infants with differing degrees of ROP and document the long-term eye status and visual function of very low birth weight infants with varying degrees of treated or untreated ROP. Three specific questions will be examined: 1. Are there long-term structural or functional ocular sequelae in 17 eyes treated with cryotherapy that would require re- evaluation of the risk/benefit ratio of recommending cryotherapy? 2. Do long-term data on structural and functional sequelae in 283 untreated eyes with ROP indicate that the "threshold" for cryotherapy should be lowered? 3. Is the incidence of structural or functional ocular abnormalities any different in 98 infants with mild ROP than in 63 infants with no ROP? If not, future resources could be focused on follow-up of infants who have moderate or severe ROP.